Orbit Post Sitemap
ASML's latest report is not bad. Net sales in the third quarter reached €11.7 billion, with net profit around €5 billion; the company also raised its full-year 2026 sales forecast to between €44 billion and €45 billion. More importantly, the equipment demand driven by AI remains strong, with some customers even inquiring about orders for 2029 and 2030. But here is a question worth considering: if AI demand is so strong, why are chip stocks still pulling back due to market concerns? I believe the core reasons lie in three aspects. 1️⃣ AI demand is real, but stock prices may have already priced in expectations in advance. The market is buying not only today's performance but also growth for the next few years. If investors had overly high expectations before, even good earnings reports might lead to profit-taking. 2️⃣ Strong orders for ASML do not mean all chip companies will benefit. ASML is part of the semiconductor equipment supply chain. Strong equipment demand indicates the industry is still expanding, but whether this translates into profits for chip companies depends on order fulfillment, capacity utilization, and end-user demand. 3️⃣ TSMC is the next key validation point. The focus will be on TSMC's statements regarding AI chip demand, advanced packaging, and capital expenditures. If order growth and future guidance remain strong, market confidence in the AI industry chain may be supported; if guidance falls short of expectations, valuation adjustments should be watched for. Some of my views: Currently, we cannot directly conclude from the chip stock pullback that... Gemini 4 Argon 正式发布，主攻长程推理、企业网络安全和 agentic 编程，直接对标 Claude Opus 5.5 和 GPT-6 Astra。定价是每百万 token 输入 2 美元、输出 10 美元，谷歌把前沿级推理拉到了 Sonnet 同档。
这一步比能力榜单上的位次变化更值得留意。同等能力砍到中档价位，用户的选择逻辑会变：不再被预算卡在「要不要用最强模型」上，而是转向响应速度、生态完整度和产品细节。竞品的定价权通常就是这样一点一点被削掉的。
Claude 这边的压力在于怎么接这一拳。被拉到同价位后，要么跟价压缩利润，要么拿出更硬的差异化证明自己贵得有理由。领先本身不构成护城河，能把领先卖出溢价才算。能被对标的产品，迟早要被定价。昨天才说的 $btw 没多少人玩
结果今天继续拉到了2.6，爆了300多万空单了
现在mm又有钱继续拉了
这种极度高控的币为啥非要去空呢？ 今晚真正值得盯的事件可能不是9月CPI，而是Anthropic的闭门会议。
据CNBC报道，Anthropic今晚面向核心机构投资者召开闭门会，首次公开披露第三季度完整财务数据，为其筹备中的IPO铺路。两名匿名知情人士也确认了会议安排。
这次闭门会的核心目的，是在正式递交S-1前，向资本市场传递最新营收增长与盈利指标，帮机构建立更清晰的估值参考。管理层和合作投行也会通过参会反馈，评估市场对IPO的接受度与定价预期。
真正值得琢磨的是时间顺序：正常流程是递交S-1后所有人一起看到数字，而Anthropic在公开递交前，先把最完整的一季账本单独摊给机构看。先让能定价的买家拿到数据，再让公开市场看到，中间的时间差本身就是一次定价。
Anthropic此前为了等三季度财务数据推迟了IPO节奏。能等，说明管理层对三季度的数字有信心。愿意慢一步，也要把最完整的一季账本摆上台面。
接下来要观察的是数据本身。营收增速和盈利指标会直接影响机构对估值的锚定，也会影响后续公开递交S-1时的市场反应。如果数据低于预期，闭门会上得到的反馈会先于公开市场体现出来。PancakeSwap 一项治理提案把 11 条支持链上 v3 与 Infinity 的 LP 手续费分成统一提高到 88%，协议手续费降到 12% 并注入国库，v2 不变。按 2025 年交易量估算，LP 预计每年额外获得约 6200 万美元手续费，交易者承担的费率不变，但流动性深度预计最多提升约 30%，价格影响最多降低约 20%–25%。
多给 LP 的那部分，不是凭空印出来的，是从协议自己留的 12% 里让出来的。一个协议愿意主动少赚手续费，通常说明它更怕流动性跑掉，而不是少收一点费。
同时，提案计划把每日 CAKE 排放量从 22500 枚削减至 7000 枚，降幅约 69%。这 7000 枚全部分配至生态增长基金 Pantry，每年减少铸造约 570 万枚 CAKE。CAKE 总供应量自 2025 年 4 月通过 Tokenomics 3.0 后已下降 11.5%，提案预计维持 2030 年供应量减少约 20% 的目标。一边让利给 LP 留人，一边减少增发，两件事都指向同一个目的：让这个池子对提供流动性的人更有吸引力，而不是对等着领新币的人。Today's summary in one sentence: Bitcoin traded sideways all day between 83,000 and 84,100, currently priced at 83,475, moving only +0.15% in 24 hours. The total liquidations across the network dropped from 770 million yesterday to about **150 million** — before the CPI release, the market has pulled back all leverage, waiting sideways for the 8:30 PM announcement tonight. Comparing with this morning's expectations, where was the difference? This morning I said that under a hawkish scenario, leveraged long positions would continue to be squeezed, and BTC might test 80,951 again. The actual movement was the opposite: the bulls were not hit further; the shorts were the ones squeezed. Of today's 150 million liquidations, shorts accounted for 52%, and among BTC liquidations, shorts made up 78%, with the largest single liquidation being a BTC short position worth over 3 million USD on Binance. The script is simple: before CPI, the price did not drop, the 83,000 level held steady, and those shorting couldn't hold on and were forced to buy back to close positions — this is a classic short squeeze, where shorts themselves supported the price. Why was it sideways and not a drop? Three reasons. First, yesterday's 770 million liquidation cleaned out most high-leverage longs, so today no one dared to open large positions before CPI; both longs and shorts retracted, naturally reducing volatility. Second, spot buying is still ongoing: the price briefly surged to **84,147** (today's high) in the morning session, though it didn't hold and fell back, but every dip below 83,000 met buying support, keeping the price stuck in a range between the 50-day moving average at 81,000 and the ceiling at 84,000. Third, on the macro front, neither bulls nor bears have new cards: the 10-year US Treasury yield fell from 5.36% to 5.2%.#Tokenized US Treasuries Used for Large-Scale Triparty Repo for the First Time
The boss has something to say
BNY Mellon and Castle Securities completed the first large-scale triparty repo using tokenized US Treasuries as collateral financing. Castle purchased two-year US Treasuries, BNY generated on-chain tokenized securities, settled within minutes, then used them for overnight repos. Tokenized US Treasuries have moved from issuance and settlement into institutional collateral financing scenarios. Ondo-related products have surpassed $3 billion in scale.
I believe this is an important step for RWA implementation. Previously, tokenized bonds could only be held and traded; now they can be used as collateral to borrow money. Institutional funds entering on-chain have gained a real exit. This is a long-term positive for RWA and stablecoin settlement.
But in the short term, it hasn't boosted coin prices. Bitcoin still follows macro trends. US Treasuries 30-year yield is 5.7%, ETF funds are still flowing out, and CPI is about to be released. I'm currently out of position, not catching falling knives. $BTC $ETH $ZEC
The above analysis is time-sensitive; orders must have stop-losses set. Good luck.今晚CPI其实不是重点，重点是美联储的信誉正在被市场当笑话看。哈玛克放鹰说还要加，市场却把10月按兵不动的概率干到77%。两边完全不在一个频道上。
大饼$BTC 这边，美国机构在减仓，巨鲸三周增持超八万枚。机构在撤，长期持有者在接。如果CPI偏热，大饼先被砸一波，但利空出尽后巨鲸可能加速吸筹。如果核心环比只有0.2%，大饼能试八万四到八万五。
以太$ETH 现货ETF上周净流出超五亿，贝莱德跑了近五亿。但有个细节值得看，它的杠杆率降到了七个月最低，资金费率转负，空头在付费持仓。如果价格稳住不破关键支撑，空头回补可能成为反弹燃料。前提是ETF资金得先停下来，现在还没看到。
黄金$XAUT 现在撤的是期货市场的投机多头，瑞银说这波下跌就是非商业账户在削减净多头。但黄金ETF持仓异常抗跌，资金持续流入，央行连续二十三个月增持，九月增持量还是本轮最大。投机盘在跑，配置盘在接。大饼的接盘方是链上巨鲸，黄金的接盘方是央行，资金属性和耐心完全不在一个量级。
别赌数据，赌美联储的信誉危机。#九月CPI今晚公布，美联储加息分歧待解 今晚20:30，美国9月CPI登场。别只盯着整体0.6%的油价噪音，真正的暗线是核心CPI：预期0.2%，前值0.3%。📉
市场给10月加息概率仅17%，鸽味很浓。但若核心环比冲到0.3%，这17%会像弹簧一样弹回去——因为核心剔除了能源，才看得清通胀是否扩散。美联储9月纪要也警告：能源高位若持续，可能渗透到更广物价。⚠️
所以今晚逻辑：整体高=已知，核心高=意外。核心0.2，风险偏好喘息；核心0.3，美元/美债收益率上行，$BTC 先杀杠杆再找支撑。BTC会跌到哪？先看前低/清算密集区与ETF流入是否托底，别拍脑袋。📊
你押0.2还是0.3？超预期BTC看哪个位置？一起学习🧠#九月CPI今晚公布，美联储加息分歧待解 #台积电Q3营收创新高，盈利表现待揭晓 #韩股急跌后反弹收窄，半导体买盘能否持续？ 《加息概率飙升，大饼为何涨不动》
刚看了CME美联储观察：10月按兵不动概率77.3%，加息25个基点概率22.7%。一个月前，这个加息概率还是零。12月更鹰派：加息25个基点概率68.7%，50个基点19.1%，不加息仅12.2%。市场几乎认定年底前必加息。
我认为这才是BTC涨不上去的根本原因。美联储从降息预期切换到加息预期，高Beta资产自然承压。CPI若超预期，加息概率还会继续上调。今晚数据见分晓。
$BTC $SOL $ZEC
#CPI与PPI公布在即，加息预期迎考验 #台积电Q3营收创新高，盈利表现待揭晓 #半导体板块集中抛售，韩国KOSPI跌超6% 法国议会正在审议2027年预算案，拟从2027年1月1日起，把加密货币兑换成受MiCA监管的稳定币纳入31.4%资本利得税范围。目前所有措施都尚未成为法律，只是一份还在走程序的修正案。
这条草案真正改变的不是税率，而是「什么算处置」。很多人把比特币换成USDT当作换个存放形式，不算卖。一旦把换稳定币划成应税事件，那些为了避险、为了不动用银行而临时换成稳定币的操作，每次都要先算一笔税。
Chainalysis法国区CEO Francois Volpoet警告，这个方向会促使交易者转向欧盟监管之外、难以监测的稳定币和离岸渠道。税率越高，资金越往看不清的地方走，这是可以预见的反应。
同一批措施里还包括：申报价值至少10万欧元的自托管加密钱包，以及把加密资产纳入离境税范围。单独看是税务合规，放在一起就是把自托管和出境两条路径都纳入监管视野。
财政委员会10月7日通过修正案，10月9日支持了自托管申报和离境税措施，但随后以31票对3票否决了政府2027年预算收入部分。相关加密税收修正案需要提交全体国民议会重新审议，收入部分投票预计10月20日举行。NEAR at $5, are you itching to buy?
CPI will be released in an hour and a half tonight. Long positions just dropped from a historic high of 1.92 billion, with liquidation lines at 4.96 and 4.89 right below. Are you trading or gambling?
Let's look at the surface first: from 2.3 to 6, more than doubled in a month, now slammed back to 5. Everyone is asking—can you catch the bottom?
I tell you: $5 is the middle of the range, not the starting point of a rise. Opening new positions before CPI is like catching a flying knife with your eyes closed.
First thing: this surge was driven by leverage, not spot buying.
NEAR futures positions were only 300-450 million in July and August. They exploded starting mid-September, reaching a historic high of 1.92 billion on October 12. Price dropped from 6 to 5, but positions only fell by 20%, still at 1.56 billion.
In plain language: leverage hasn't been cleaned out yet.
Binance funding rate annualized at 6-7%, Hyperliquid at 11%. Longs are still paying, but it's no longer extremely crowded. The real risk is—the first tier of long liquidations is at 4.96 and 4.89, just 1%-2% below $5. If CPI is hotter than expected, this layer will be wiped out first.
Chasing longs here with a stop loss within $0.2 is like giving a gift to the needle. Daily ATR is 0.54, daily volatility of 10%-16% is normal.
Second thing: the benefits of ETFs and Intents are already priced in.
Bitwise spot NEAR ETF, ticker NRR, listed on NYSE Arca on September 29, the first in the US. First week inflow was $58 million.
Sounds exciting? But NEAR's market cap is 6.6 billion; $58 million can't support a second wave unless inflows accelerate.
NEAR Intents cumulative volume is 31-33 billion, covering over 30 chains, with estimated annualized fees of $45 million, some of which are used for buybacks. But there was a $3.86 million exploit in early October, recovered within 24 hours. The money came back, but the ETF story is selling this product, which was just hit, so valuation shouldn't be considered risk-free.
Also, inflation dropped from 5% to 2.5%, with governance proposals aiming to reduce it further to 1.6%. This is a slow variable and won't help tonight.
Wintermute observation: funds shifted from high-liquidity DeFi and large-cap coins to AI. That week, only 11 of 66 large-cap altcoins closed up, median -3.7%, NEAR was a winner. The winner's position is that $6 needle.
Third thing: technicals show weekly bullish, daily distribution.
Weekly RSI 68, moving averages bullish, large scale intact.
Daily RSI 55, MACD histogram turned negative, price below 9-day MA. Small scale finishing this leverage cycle.
Key levels:
Resistance above: 5.37-5.60 (failed rebound means weakness) → 5.90-6.03 (previous highs supply)
Support below: 4.80 (21-day MA) → 4.62-4.64 (previous low platform) → 4.30-4.45 (real demand, last defense of the uptrend)
Trend break: daily close below 4.30, watch 4.00 50-day MA.
At $5, neither bulls nor bears have the advantage. The edge is at 4.80 pullback or a close above 5.25, not at the price you're watching now.
Bull vs bear, you decide:
On one side:
Weekly bullish, large scale intact
ETF channel open, $58 million inflow first week
Intents volume over 30 billion, real business
Inflation down from 5% to 2.5%, governance proposals to reduce further
Relative strength, outperforming the market for a month
On the other side:
Futures positions dropped from 1.92 billion historic high, leverage not cleaned
Liquidation lines at 4.96 and 4.89 right below
ETF inflows relative to market cap can't support second wave
Intents had a $3.86 million exploit
Macro: Fed rate 3.75-4%, possible hike this year, zero chance of cut
10-year US Treasury 5.1-5.2%, high beta cut first
Tonight's CPI, core data decides direction.
Consensus: overall CPI YoY 3.6%, core 2.4-2.5%.
Core MoM ≥0.3% or YoY above 2.5%: long liquidation below $5 triggers first, 4.96 and 4.89 liquidated, direct drop to 4.80 or even 4.62.
Core MoM ≤0.2% and YoY steady at 2.4%: only then can we talk about returning to 5.50.
Don't use "it's stronger than the market" to hold through CPI. Altcoins have higher beta than ETH; with the same data, NEAR is more elastic.
Trading strategy
Long in two tiers, don't catch the flying knife in the middle at $5:
After soft data, 4-hour close above 5.25, pullback to 5.10 without breaking, enter again. Stop loss 4.88. First target 5.55, second 5.90. If daily close doesn't reclaim 5.60, exit at 5.55.
Better position is 4.80-4.85, need to see lower wick and close above 4.95. Stop loss 4.60. Use half normal position size because holdings are still high. Target first 5.40.
Deeper plan orders at 4.35-4.50, stop loss 4.15. This is a buy point while this rally is still alive. If daily close breaks 4.30, plan invalid, next support 4.00, not "add another tier."
Short only two types:
Rebound 5.40-5.60 fails, 4-hour close can't reclaim. Stop loss 5.70. Target 4.90, then 4.64.
4-hour close below 4.88, rebound 5.00 fails. Stop loss 5.20. Target 4.64, extend to 4.35. If 4.90 holds, don't chase shorts, risk-reward is poor.
Structure judgment fixed:
Daily close below 4.30 ends the uptrend since September, only shorts or no position left.
Daily close above 5.60 ends pullback, close all shorts, next stop 6.03.
6.03 without volume breakout remains resistance, not a breakout.
NEAR now is like SOL in 2021—
From $2 to $6, everyone shouted "the next Ethereum."
Then retraced 50%, everyone called it a "scam."
Then it rose to $20.
But this time, are you sure you can withstand that 50% retracement in the middle?
A bull market isn't without pullbacks, but every pullback feels like a bear market. $BTC $ETH $NEAR 美国将于北京时间10月14日20:30公布9月CPI。市场预计能源价格上涨可能推动整体CPI同比由3.4%升至3.6%—3.7%，核心CPI环比增速则可能由0.3%放缓至0.2%。近期哈玛克、威廉姆斯及柯林斯相继释放年内可能继续加息的信号，但未明确主张10月立即行动。CME FedWatch数据显示，市场对10月维持利率不变的定价约77%，加息25个基点的概率约23%。当前分歧集中于加息时点：若核心通胀高于预期，10月加息概率可能回升；若价格压力主要来自能源，暂停预期或得到强化。数据结果将进一步影响美债、美元、黄金及BTC等资产定价。📉 Today's market feels like a script change
$ZEC followed BTC down to around 1120 in the early session, but stubbornly pulled back to 1220 by midday, with bears getting slapped back and forth. 🧨 Multi-timeframe indicators still lean bearish, open interest has clearly shrunk, and the long-short ratio is 46:54: more shorts with light positions; fewer longs with heavy positions. The capital structure is quite conflicted. I tend to think this V-shaped rebound is a dead cat bounce, a struggle during the downtrend, not a reversal. 🐱
$NEAR is best avoided for now. Thin liquidity makes the order book easily manipulated; it's easy to get in but not necessarily graceful to get out. ⚠️
Two most painful regrets in trading: not taking profits and watching them evaporate; or getting scared and selling early, missing the main rally. The root cause is often not setting the timeframe before entering a position. Entering on a 15-minute chart but expecting a 4-hour trend, then clinging to the trade after the move ends, turning unrealized gains into stop losses; entering with a 4-hour logic, but getting shaken out before the trend breaks, only to watch it run further. 😮💨
Short-term trades capture short-term gains, long-term holds endure long-term volatility. When timeframes get mixed up, mindset collapses. Set your timeframe first, then talk direction. 🧭 #SeptemberCPI announced tonight, Fed rate hike disagreements await resolution #TSMC Q3 revenue hits record high, profit performance to be revealed The market is stuck halfway up the hill, with neither bulls nor bears daring to make the first move. Fed's Williams says there's no rush to raise rates, no worries before 2027, risk appetite has slightly rebounded, ZEC leads the rebound, and South Korean chip stocks have stopped their five-day losing streak. But the chip sell-off isn't over yet, tech stocks have cooled off again, and Bitcoin just touched 84,000 but was pushed back within an hour. Everyone is now waiting for the September CPI; without the data, no one dares to go all in.
BTC is currently at 83,412, up 0.2% in 24 hours. Resistance at 84,000 is strong, short-term support is at 83,000, and if broken, look to 82,000. Before the data, expect narrow-range grinding; don't expect it to pick a direction on its own.
ETH is currently at 2,503, up 0.28% in 24 hours. It keeps fluctuating around 2,500, down 4.36% over seven days, much weaker than Bitcoin. 2,480 is the bottom line; if it can't hold above 2,520, no talk of a rebound.
ZEC is currently at 1,239, up 3.35% in 24 hours. It dropped 80 dollars from 1,282, a quick rise and fall just played out; those chasing highs have already been left behind. It is the vanguard of the rebound and also the first wave of the sell-off.
Don't get itchy before CPI lands. Jumping in now isn't bottom-fishing; it's feeding the data as fuel.
$BTC $ZEC $ETH
#韩股急跌后反弹收窄，半导体买盘能否持续？ #台积电Q3营收创新高，盈利表现待揭晓 #九月CPI今晚公布，美联储加息分歧待解 This whale closed out $15 million worth of long positions at breakeven before the CPI release. Could the CPI tonight be a bombshell?
The whale with address starting 0x4ff9 completely liquidated about 6,000 ETH long positions (approximately $14.98 million) in the past hour, exiting near breakeven and clearing the entire account.
With the CPI data about to be released, this whale is closing positions near breakeven, showing a lack of confidence in the upcoming market rally?
Currently, $ETH is oscillating narrowly between 2485-2510, with the price basically hugging the 2500 whole number level.
The short-term structure is weak, price is below multiple short-term moving averages, RSI is neutral to weak, and volume has not increased, overall in a compressed state before the data.
Support levels:
First support: 2470-2485 (near intraday lows, losing this would weaken the short-term structure)
Key support: 2450-2430 (previous low concentration area, a decisive break could accelerate downside)
Stronger support: 2400-2410 (recent rebound starting point, breaking this would turn short-term bearish)
Resistance is seen at 2520-2530, then further up at 2550-2560.
The whale choosing to close at breakeven before CPI is more about avoiding data uncertainty rather than confirming a direction.
After the data release, the focus is on whether 2470 can hold and if funds re-enter the market. What’s your take on tonight’s volatility? #SeptemberCPI announced tonight, Fed rate hike divergence awaits resolution The US September CPI will be released tonight at 8:30 PM. The market expects overall inflation to jump from 3.4% to 3.6%-3.7%, mainly driven by energy, with gasoline prices rising more than 7% in one month. However, the core CPI is actually cooling down, with the month-on-month expectation dropping from 0.3% to 0.2%, and the year-on-year around 2.5%, indicating that price increases are basically concentrated in the energy sector and have not spread to other areas.
Recently, Williams and Collins have consecutively taken a hawkish stance, both saying there might be another rate hike within the year, but no one advocates for action in October. The probability of a rate hike in October is currently only about 23%, with the market betting on December. What can truly change this pricing is the core CPI. If the core month-on-month stays within 0.2%, the expectation of a pause will be more stable; if the core also exceeds expectations, then the probability of an October rate hike will have to be revised upward.
$BTC is now around 83,000, rebounding from just above 80,000. There likely won't be major moves before the data release tonight; the key is how US Treasury yields move afterward. If the core CPI holds steady and Treasury yields ease, Bitcoin will have a chance to test 85,000 again. If the core exceeds expectations, the 82,000 support will have to be tested again. #九月CPI今晚公布，美联储加息分歧待解 $SOL 这条链上有一类钱不炒币，是现实世界的资产搬上来做生意的。
按数据站的分栏，这一类现在一共5.85亿美元，分在12个项目里。最大的两个，一个做再保险，2.93亿，一个做信贷，2.49亿，两个加起来占了这一类的九成二。
剩下的都是小个子，紧随其后的一个3360万美元，再往下是几百万甚至几十万，这一类的规模基本由前两名撑着。
这类资产背后是现实世界的现金流，保险的赔付和贷款的利息，搬到链上来图的是结算快和账目透明，吃的是现实现金流那口饭，币价涨跌它们不关心。
链上的借贷和做市那两类，规模都是二三十亿美元，这一类还在个位数亿。放在SOL的638亿美元市值旁边，不到百分之一。
SOL的市值动一个点，是6.38亿美元，比这一整类还多，所以这类钱现在进出都影响不到币价，价格的涨跌还是币自己的事。
但它的方向值得看。这类钱进得快，背后是有机构愿意把现实资产放到这条链上结算，而不是只在上面做交易。
数据站上这类资产有单独一栏，隔一段时间看一次，能看出这类钱进得快不快。$ZEC is stuck in a frustrating sideways range ahead of tonight’s CPI data. ⚡
Both sides are waiting for a mistake. Once the data hits, one side could get liquidated as price breaks out.
#ASMLQ3ResultsOct14 特拉华州一纸注册：HYPE的ETF走到哪一步了，我把流程拆开给你看
先说事：公开报道显示，一个以$HYPE 为标的的 ETF 信托实体，1月8号在特拉华州完成注册，1月9号相关报道发出来。看到"ETF"三个字先别激动，我把整个流程拆开说。
一个 ETF 从想法到落地，要经过实体注册、正式申请、监管获批、上线运行好几道关。现在走完的是第一道，只是"有人把公司注册好了"，离获批远着呢，更谈不上资金流入。那这事重不重要？重要。它说明有机构愿意为 HYPE 铺路，机构化的想象空间被打开了一条缝。但也仅此而已——市场总喜欢提前交易预期，预期跑在文件前面，价格波动往往就是这么来的。
我的做法：把这个进展记在心里，眼睛盯后续的正式文件。想象力归想象力，盖章才算数。信仰不变，等文件。Anthropic's decision to show Q3 performance privately to institutional investors makes the reported $2T valuation less a verdict than a negotiating reference point. Demand can be tested before full earnings face public scrutiny, while the offering size and timetable remain open.
That preserves flexibility, but it also leaves outsiders with a headline valuation and few inputs for judging it.
#AnthropicIPOValuation #Strive增持3311枚BTC，高息融资能否持续？
帮主有话说
Strive又加仓了。10月5日到9日，斥资2.8亿美元买了3311发枚BTC500，总持仓到32773枚。同时打算万股SATS优先股，每股100美元，年化股息15%，募资继续买币。
Strategy反着来。过去两周花了3.69亿美元回购STRC优先股，只增持584枚BTC。一个加杠杆买币，一个降杠杆保现金流。
我的判断是，高息融资是把双刃剑。牛市放大收益，熊市15%的股息就是硬支出。币价一波动，现金流压力就上来。Strive赌的是BTC继续涨，但一旦横盘或下跌，融资成本会吃掉利润。
市场关注点已经从持币规模，转到融资效率和资本结构。谁的模式更稳，谁活得久。
我现在空仓，82800的多单今天止损了。CPI出来之前不接飞刀。ETF资金还在流出，美债30年期5.7%，宏观压力没消。财库增持是慢变量，改变不了短线节奏。$BTC $ETH $ZEC
不追涨不杀跌，等数据落地再找位置。
以上分析都具有时效性，单子必须要挂好止损，祝君好运。【Crypto Script】
#September CPI announced tonight, Fed rate hike divergence to be resolved
I'm Script Bro, the CPI is coming again tonight
The crypto world is probably facing another big guess-the-size event
The market's biggest dilemma now isn't whether inflation is high or not
But whether the Fed still wants to raise rates
To put it simply, the US economy is a bit awkward right now
Employment is cooling down, but inflation hasn't fully softened yet
On one side, it's getting harder for people to find jobs
On the other, prices just won't come down
The Fed is stuck in the middle, getting slapped from both sides
But I think people might be overlooking one issue
The market isn't really trading the CPI itself
It's trading the gap between CPI and expectations
No matter how bad the data is, as long as it's not as bad as expected, it can still go up
No matter how good the data is, if the market has already priced in too much
The good news might turn into bad news when realized
That's why in crypto we often see prices surge first when news breaks
Then crash afterwards
It's not that the big players can predict the future, but that leverage positions are too crowded
Especially now BTC itself is caught between macro pressure and institutional allocation tug-of-war $BTC $ETH $ZEC 今晚CPI，核心才是那根针
兄弟们晚上好。
20:30，美国9月CPI。整体预期环比+0.6%，油价推的；核心预期从0.3%降到0.2%。10月加息概率已压到17%。但只要核心偏离0.2%，比如还是0.3%，加息预期立刻翻脸。
逻辑：整体高不可怕，核心高才要命。核心剔除了能源，看的是通胀扩散没扩散。9月纪要也警告，能源高位可能传导到更广物价。
我的仓位：BTC多单浮盈35%，ETH多单浮亏40%。浮亏在收窄，但今晚才是大考。设好底线，等数据。
你赌核心0.2还是0.3？若超预期，BTC会跌到哪？依据是什么？一起学习。
$BTC $ETH $ZEC
#九月CPI今晚公布，美联储加息分歧待解
#BTC现货ETF再度净流出，机构买盘能否回暖？
#交易之声：你的经验值得被听到 $BTC 84000被摁了两次，今晚CPI能不能给个痛快？
BTC现在卡在83385，一天晃下来几乎没动，就跌了0.42%。好几次摸到84000的边，转头就被硬生生打回来，上也上不去下也下不来，所有盯盘的人都悬着一口气等靴子落地，这种磨人的横盘滋味，最近守盘的老玩家肯定都懂。
资金面刚冒了点暖意，昨天现货ETF净流入1.76亿美元，是近两周最多的一天。但10月第一周整体还是净流出3.86亿，单日的回流说明不了什么，更像是大资金在数据公布前悄悄把仓位摆好，就等数据出来的瞬间冲出去。
别被表面的同比3.6%预期骗了，这个数字是能源涨价硬推上去的，美联储根本不拿它当参考，核心CPI才是今晚的胜负手。现在机构两边完全反着来，高盛预判核心环比0.24%偏强，花旗却只给0.18%，说通信服务价格会拖后腿。
今晚无非两条路：核心数据低于0.2%，加息预期松绑，BTC冲84000站稳看85000；数据超预期偏强，加息话题重提，BTC直接回踩81600，破位就往80000走。这种关键节点没人想踏空或被甩下车，盯着核心数据和资金流向，比瞎猜涨跌靠谱得多。
#九月CPI今晚公布，美联储加息分歧待解 仅剩90分钟！今晚CPI，决定全球流动性走向
如果通胀依旧维持高位，降息预期会再度下调，美债率先反应，黄金、纳指的定价逻辑全部重新洗牌。倘若数据意外走弱，市场会押注暂停加息甚至转向宽松，风险资产压力得到释放。核心CPI才是关键底牌，租金、服务业价格与薪资这三项，直接决定美联储后续的政策动向。
眼下全球盘面集体进入蛰伏状态，波动被刻意压制，交易员大多减仓观望，静等数据指引。CPI落地前不要轻信任何突破行情，大多都是诱单。
今晚八点半那根瞬时K线，才是真实行情的开端：数据低于预期，行情率先冲高；高于预期则美元走强，承压下行；如果刚好符合预期最磨人，大概率插针洗盘之后再选择方向。今晚我们看的不只是K线，更是全球流动性的开关。还有一个半小时，不要眨眼。$BTC $ETH #九月CPI今晚公布，美联储加息分歧待解 84360 is not a benchmark, it's the short sellers' sniper line
At 8:14 AM, $814 million liquidated, 85% of which were long positions. Many think the bloodbath is over and a rebound is due. In my view, this rebound is not stabilization but a downward continuation giving shorts some points.
The market is very honest. Bitcoin tried to leverage the Houthi attack on Saudi Arabia and Brent crude breaking 106, but by evening it barely touched 84,000 and couldn't hold, falling back to 83,500; the altcoins didn't even follow, ETH is stuck below 2,500, SOL only at 107. Market dominance rose above 60%, which looks like Bitcoin is strong, but actually the whole market is shrinking in volume with funds huddling together for warmth—this solo run usually ends with Bitcoin catching down.
The macro environment doesn't support bulls either. Oil prices surged, making it hard for tonight's 8:30 PM CPI to disappoint; overall year-on-year is heading for 3.6%-3.7%, 10-year US Treasury yields at 5.24% are suppressing, and the probability of a rate hike in December has reached 70%. Core month-on-month ≥0.3% means repricing for hikes; if it hits 0.4%, tonight will be the bulls' second liquidation.
So I don't treat those three levels as benchmarks at all:
$BTC 84,360, $ETH 2,580, $SOL 113 are three walls; if prices bounce up to these walls, it's shorts adding positions. If Bitcoin breaks 82,200, 80,000 is next; altcoins have already fallen out of respect, Bitcoin's catch-down will only be delayed, not absent.
Don't chase the first spike when data lands, wait for it to bounce back to the wall on its own.
#September CPI announced tonight, Fed rate hike divergence awaits resolution $BLUR , $YGG , and $SAND were once hot gaming and NFT tokens, but teams have gradually cashed out and unlocked tokens. Many still hope for a return to previous highs, but gaming and NFTs remain among crypto’s weakest sectors, especially with poor game quality and fierce competition from online, mobile, and AI games.我是中线情报哥。
今日盯盘，有个逆势空头得重点关注。
市场普遍押注10月暂停加息，但交易员“greater-fools”仍在加码$BTC 下跌赌注。昨日至今他投入约4381美元：预测市场“年底跌至6.5万”持仓13.61万份（隐含概率22.5%，价值3.06万）；“年底跌至7万”持仓5.58万份（概率35.5%，价值1.98万）。
更狠的是Hyperliquid，他仍持26.97枚BTC空单（均价69962，价值225万），目前浮亏36.52万，s扛不止。
今晚美国9月CPI是关键。若通胀超预期，市场重定价紧缩，或成BTC短线压力源。
中线看，机构抛压未止，这位置多空博弈极致，别盲目追，守住筹码等方向。
$ETH
$SNDK
#九月CPI今晚公布，美联储加息分歧待解 贝莱德把美股从超配降到中性，转去偏好短债，这个动作比“看空美股”四个字细得多。
最新周评里，它把短期美债提高到超配、长期美债提高到中性，同时仍然偏好美股里的AI相关公司。它自己将这次调整描述为短期风险管理，保留对AI基本面的信心。长期方向没变，眼前愿意付的风险代价变了。
我觉得这比天天争论AI有没有泡沫更值得琢磨。同一家好公司，当债券给出的回报更有吸引力时，股票也得提供足够的补偿。你可以相信未来增长，也可以暂时降低总体权益风险，两者没有必要互相否定。
尤其这次贝莱德偏好短债，说明它并没有简单押注长端利率马上大跌。选择期限和选择股债比例，是两个需要分别考虑的问题。把增加美债配置统统解释成看好经济衰退，也会漏掉它实际说的理由。
昨天支持AI，今天降低配置，就算不坚定吗？市场很喜欢给机构贴上多头、空头的标签。我反而更欣赏把价格和回报一起重新核算的做法。
当然，这份观点不能代表它所有基金都会同样交易。读完之后，我更想问的是：一个方向再好，现在的价格还给投资者留下多少余地？
#贝莱德转向超配短债，下调美股评级 通胀数据公布前，总结一下几个比较关注的资产近况及爆发的前提。如果高利率继续维持，就看个别加密资产的结构变化。
从盘面上看大都呈现区间震荡，这个月可以关注一下NEAR、UNI、SOL和FIL这些资产的结构事件变化带来的趋势行情。
当然，如果重回降息预期，就皆大欢喜了~
$BTC $ETH $SOL 今天的盘面，用四个字形容——内忧外患。 比特币今天经历了剧烈的双向波动。 亚洲时段，受韩国KOSPI指数暴跌5.5%、较6月高点回落38%的拖累，比特币一度跌破81,000美元，最低触及81,300美元附近。 随后买盘强力介入，价格反弹至84,000美元上方，日内涨幅一度达1.55%。 但欧洲时段再度承压回落，截至发稿BTC报约83,000美元下方，24小时下跌约0.53%。 以太坊同步承压，一度跌至2,400美元，现报约2,413美元。 过去24小时，全网爆仓金额高达8.03亿美元，共有17.21万名交易者被清算，其中多头占6.94亿美元，比例达86%。 比特币和以太坊期货成交量达到2,120亿美元，较前一日的1,260亿美元增长66%； 未平仓合约名义价值为1,434亿美元，下降1.58%； 清算金额飙升284%，至8.13亿美元。 高成交、未平仓合约收缩、强制平仓增加——这不是新增做多，这是杠杆被强行清洗。 恐惧与贪婪指数从一周前的67回落至56，仍处于“贪婪”区间，但降温明显。 1， 过去三天，比特币走了一波探底-冲高-回落的完整结构。 10月12日，亚洲芯片股遭遇大规模抛售。$BTC currently has a particularly asymmetric situation between longs and shorts: the liquidation exposure amounts on both sides are almost the same, but the number of positions differs by three times.
The short risk is spread across more than 480 positions, while the longs only have 150. With the same amount, one side consists of densely packed small retail traders, and the other side is a few large holders with heavy positions.
The longs are backed by big players, the shorts by scattered retail traders. The most profitable group of addresses has just shifted from bearish to bullish, and the funding rate has flipped from a rare negative to positive — these three things all point in the same direction.
More importantly, above 87,000 there is a cluster of short stop-loss orders totaling tens of millions, just five points above the current price. If the price moves up a bit, these retail shorts will turn into buying pressure themselves; the fuel is already prepared.
But one awkward fact must be acknowledged: the formation looks good, but incremental demand hasn't arrived. Apparent Demand is still underwater, and on-chain unrealized loss supply has just fallen below 30%, indicating selling pressure exhaustion, not demand activation.
84,000 has been the repeated resistance wall these past weeks; only when it breaks above 87,000 will that cluster of shorts start to panic; below, 81,000 is the next support, and if broken, look toward 80,000. $WAL at this position, I'll be honest, don't fomo.
The Sui Foundation invested 30 million dollars into DATA Network, and Walrus aims to be the default storage for its AI applications. The AI storage narrative really has some solid backing, it's not just empty talk. It surged 35% in 24h, nearly 50% in 7 days, with clear volume increase—real money is coming in.
But precisely because it has risen like this, I want to advise caution—there's no reference point above, chasing it is just gambling on whether there's another wave of sentiment.
I've seen sector rotations many times. The storage sector has indeed been bleeding these past few days, and the old $FIL holders are still lying low. Funds are withdrawing from old storage to pour into new narratives; I've seen this happen once, twice, three times. Walrus is worth watching, but chasing at a high level means you're not betting on fundamentals, you're betting that someone else will catch you without blinking. Just a word of advice, then walk away. $FIL 三分钟落子，抵得上传统清算体系三天的长考——这不是快棋，是有人把整盘棋的计时器直接拆了。
国际象棋里最危险的从来不是对手的进攻，而是他悄悄完成了王车易位，并且把车摆上了唯一那条开放线。十月十三日发生的事就属于后者：托管方把做市商手里的两年期美国国债搬上私有账本，隔夜回购融资还没开价，抵押品已经结算完毕。这一步在开局谱里找不到，却把整个融资中局的子力活动性重新定义了。
要看懂这局，先看清棋盘上的三方：托管者是王，做市商是后，融资方是那条必须不断推进的兵链。过去抵押品一旦入柜就等于被牵制住的重子，动不了、换不掉、反复质押还要层层确认。现在它变成了能反复出动的轻子，同一枚子在一夜之间可以参与数次战术交换，而每一次交换都在压低资金成本。这就是子力效率的胜利，不是速度的胜利——特级大师从不追求快，只追求同一时间内多出来的那一步。
真正的杀招藏在第二步。挂钩代币化债券指数的总收益互换被搬上另一条链，等于把半开放线接通了；代币化国债产品规模越过三十亿那条线，则说明兵已经推到第七横线，只差升变。这些不是三个孤立的战术组合，而是一条完整的进攻谱。
但局面性隐患同样清晰。跨平台流动性是一张被切碎的棋盘，合规是双重牵制——你不能同时保护被钉住的两枚子；风控则是随时可能到来的长将，稍有不慎就是无限循环的将军与被将军。当抵押品在多个账本之间连环再质押，链条拉得越长，越像一串连续将军的战术组合：精彩、紧凑、毫无容错。任何一步不精确，立刻陷入无子可动的窒息局面。
成本能否真正下降，本质上不是技术问题，而是权威棋谱归属问题。谁的账本被认作终局记录，谁的结算时钟更短，谁就握有主动权与先手。把股权类资产也放上同一张棋盘之后，跨市场的攻击将从单线突破变成双重攻击，防线要同时应对两条线路的压力。
这盘棋才走到开局第十五步，中局的复杂性远未展开。谁先把抵押品的活动性养成一枚通路兵，谁就把所有人拖进了自己版本的残局——而对面，连钟都还没按下去。 #tokenizedtreasuryrepoMonitoring data shows that one week after ENA unlock, related insiders have cumulatively transferred tokens worth nearly $50 million to exchanges. The unlocked tokens are beginning to enter trading venues, and the potential supply pressure is worth noting, but transfers do not yet confirm sales. Next, watch whether these addresses continue to deposit and if there is actual evidence of selling to determine if these tokens are being cashed out.
$ENA 美光的承重柱在一夜之间出现了3.88%的沉降，闪迪那面幕墙直接开裂5.97%——这不是普通的市场回调，这是整栋存储大厦的结构性共振。三星和SK海力士的地基也在同步发出异响，所有人都听见了，但没人愿意承认图纸上那条应力裂缝早就在扩张。
我干这行二十年，最怕听到甲方说"需求旺、利润高、再往上加三层"。AI算力这波需求就像在原有地基上疯狂加盖，内存价格和利润率确实被顶了上去，可你回头看看那块地——供给端的地质勘探报告正在被重新审读。CXMT递上来一份4F² DRAM的新蓝本，说年底要切入服务器内存这条主干道，问题是：设计图很漂亮，但从图纸到浇筑完成再到通过验收，中间隔着一条没人走过的施工缝。量产这件事，一天没封顶交付，就一天不能算数。
再看原来的两家总包——三星和SK海力士的回购工程已经接近收尾，这意味着支撑股价的那几根临时脚手架要被拆掉了。外资在撤场，前期涨幅堆积的荷载还没卸载，原本看似稳固的楼体自然要往下沉一沉。抛压不是意外，是结构自重的一次释放。
真正让所有结构工程师夜里睡不着的是2027到2028年那批新增产能——那可是全新浇筑的一整片楼群。如果需求端那台AI发动机继续满负荷运转，供需两端就能维持动态平衡，剪力墙咬得住；可一旦AI的边际买盘松动，利润率撑不住估值，这批新楼就会变成空置的钢筋水泥森林，风向一变，整片天际线都会倾斜。
所以问题从来不是"还能不能涨"，而是这栋楼的核心筒到底是钢筋混凝土还是石膏板。
我看过太多项目，外立面光鲜得能上杂志封面，掀开吊顶一看，管线全是临时拼接。存储这轮行情的真实抗力，不在于新闻稿里的产能规划，不在于回购公告的金额大小，而在于那条最底层的晶圆产线能不能在良率、成本、客户验证三根主梁上同时咬合。CXMT那4F²的结构选型如果站得住，就是在别人的承重体系旁边重新打桩；如果只是概念浇筑，那2027年这片新楼群一进场，整个区域的租金都会被稀释。
AI是不是那个永不沉降的持力层？现在谁画的剖面图都不算数。得等2026年那批施工缝彻底闭合、实测沉降数据出来，才知道这栋楼是能继续往上加，还是得停下来做结构加固。
在那之前，所有估值都只是效果图。 #sandiskmicronslide#Strive增持3311枚BTC，高息融资能否持续？ $BTC
一周买 3,311 枚，均价从约 7.95 万抬到约 8.45 万。
▪️ 10/5–10/9 购入 3,311 枚 BTC，均价约 84,500 美元
▪️ 持仓由 10/2 的 29,462 枚增至 32,773 枚
▪️ 上一次披露（9/14–9/18）为 1,355 枚，均价 79,475 美元
▪️ 同期 2.08 亿美元认股权证已被行使，约 1,530 万份尚余、10/13 到期
▪️ 100 美元最低发行价已成为治理文件中的约束性条款
▪️ 单周买入量放大约 1.4 倍，均价同步上移约 6%
分歧不在买了多少，在这 3,311 枚是用什么成本换来的。
量在放大，价也在抬高——认股权证被行使带来的是股本，不是经营现金流。把行权、到期和最低发行价写进治理文件，说明这轮扩张靠的是资本工具。融资成本一旦上行，增持节奏会先于币价反应。
方向给观望 —— 权证到期后仍能维持增持才算融资顺畅；否则节奏先慢下来。#Goldman Sachs Q3 Earnings Exceed Expectations, US Bank Earnings Show Divergence
In the same earnings report, equity trading +72%, fixed income −1%.
▪️ Q3 net revenue $18.68 billion, up 23.1% year-over-year, $1.79 billion above consensus
▪️ Equity trading revenue $6.44 billion, up 72% year-over-year; last quarter $7.42 billion was a record high
▪️ Net profit $5.294 billion, up 29% year-over-year; diluted EPS $17.01, consensus $13.02
▪️ Fixed income, FX, and commodities revenue down nearly 1% year-over-year, below expectations
▪️ After earnings, pre-market up over 3%, early trading up nearly 4%
▪️ CEO Solomon warns that AI and data center financing demand is huge, but not every project will succeed
The divergence is not about beating expectations or not, but that two directions appear simultaneously in the same earnings report.
Equity trading +72%, fixed income −1%, these two figures come from different corners of the same balance sheet: the former profits from volatility, the latter from spreads and inventory. When interest rates are high, equity volatility increases while bond market trading thins out, causing the strengths of a full-service investment bank to shift.
The direction is to wait and see — fixed income turning positive next quarter would mean a full recovery; continued divergence means volatility remains the strength.把热闹拆成订单
$TIA 的生意，我会先看开发者到底花了多少钱。 项目使用Celestia发布数据，需要支付以TIA计价的费用，费用与交易和数据大小有关。 这条需求比较直接，但数据量增加多少，收入就一定同比增加多少吗？还得看实际收费。 如果容量扩大、单位费用下降，用得更多也可能只带来有限的收入增长。 所以我更想把付费数据量、单位收费、总费用放在一起看。接入项目数量很亮眼，也得有订单金额来支撑判断。
$LINK 值得研究的是收入转换这一步。 官方介绍，企业业务收入和链上服务费用，可以通过支付转换机制积累成LINK储备。 我会追踪储备实际增加多少，以及增加得是否持续。 储备积累和销毁要分开算，代币进了储备，总供给并不会因此自动减少。机制有了，实际执行规模才决定经济意义有多大。
$JTO 这里先把两种权利分清。 JitoSOL承载质押资产及相关收益，JTO承担网络治理功能。 研究质押业务增长时，我会继续查收入分配和治理提案，而不是直接把质押收益写进JTO持有人的账里。 后面真正值得关注的，是哪些规则落地以后，能给这个代币形成持续需求。 ETH is stuck at 2503, repeatedly grinding along the middle Bollinger Band. The MACD histogram just turned green, and RSI has already hit the overbought zone. This kind of short-term impulsive move diverging from overbought conditions is most prone to a wick liquidation.
Around 2501.9, there is a large accumulation of long liquidations. The price is hugging the upper edge, and as long as it breaks below 2496, the first wave of forced liquidations will be triggered, instantly opening liquidity below. When the red light came on, I glanced again at the liquidation chart. In these dense long-short liquidation zones, the main players like to trap both sides.
There are also considerable short liquidations from 2550 to 2600 above, but chasing shorts at the current price is likely to get stopped out first. So I neither chase longs nor short at the current price. I try shorting in batches between 2508 and 2516, with a stop loss at 2541 and take profit initially at 2472, pushing down to 2440 if broken.
If the 1-hour close holds above 2545, all short positions should be closed unconditionally and reversed to short-term longs, targeting 2585 with a stop loss at 2520.
$ETH
#ASML Earnings Preview: EUV Orders Are the Key Focus
@OKX Planet $SOL This 100x short SOL position's unrealized profit has skyrocketed to 515.66%, with the main bearish phase completely consumed, resulting in an extremely impressive profit cushion.
Looking back, entry was at 112.67. After high-level selling pressure was released, the bearish trend smoothly pushed down. The current mark price is 106.86. At zero-level grading, under 100x extremely high leverage, it directly exploded to over 5.1x returns, with very fierce downward momentum.
Currently, 106.86 is approaching the 106.5 and 105 integer support levels and previous low support zones. Passive buying at low levels will inevitably be dense, and volume may shrink at any time. With 100x extremely high leverage combined with over 500% huge unrealized profits, although the profit is very substantial, SOL as a highly volatile popular asset experiences especially intense order book fluctuations (often accompanied by sharp manipulations from major players). Even a slight reverse pullback can instantly wipe out most of the profits or even trigger liquidation. The tail-end cost-performance ratio has dropped to rock bottom.
Following the trend is easiest. At the current price, it is strongly recommended to close out over 95% (or even fully close) to lock in profits, leaving a very small residual position with stop loss directly nailed to the cost line (or significantly raised breakeven line to ensure absolute profit safety). Keep a close watch on the 105 and 100 levels. If volume breaks through, a very small residual position can be left to float; if volume shrinks and price stagnates or any upward spike occurs, exit fully. Protecting real cash is the key; extremely high leverage only eats the body of the fish, not the tail. $BTC $ETH #September CPI announced tonight, Federal Reserve rate hike divergence pending resolution On October 7, the yield on 30-year U.S. Treasury bonds briefly rose to 5.706%, hitting a new high since 2002. Many people's first reaction was: with such a high yield, is it time to buy long-term bonds? Basman’s reminder goes the other way—if the 20-year yield breaks 6%, some institutions might actually accelerate selling.
The issue lies in the delivery rules of Treasury futures. Sellers can choose from a basket of Treasuries the one with the lowest delivery cost after conversion, and the conversion benchmark is set at 6%. Once market rates continue to rise, the most suitable bond for delivery will switch to a longer-term, more interest rate-sensitive one.
Basman’s report on October 13 calculated that the duration of the relevant delivery bonds rose from about 11 to 13.24, and after further switching, it could exceed 17. Without increasing positions, the same interest rate volatility causes amplified gains and losses. For institutions with risk limits, this is enough to trigger position reductions or increased hedging.
Leverage further amplifies the process. Many funds borrow money to buy Treasury cash bonds while simultaneously selling futures to earn the spread. When volatility spikes, margin requirements rise or financing becomes difficult, they can only sell cash bonds and buy back futures to reduce exposure. The original Treasury buying demand directly turns into selling pressure.
Therefore, the higher the yield, the more attractive long-term allocation may be, but short-term prices depend on how concentrated the selling is and whether buyers have enough funds to absorb it. Institutions can remain optimistic about long bonds but may sell first due to risk limits and cash needs. Silicon Valley Bank fell into the same trap initially.The key test this week, the “CPI,” is about to be released, marking an important turning point for the financial markets this week. How should the data be interpreted? How should the US stock market, BTC, and bond market be priced normally? According to current expectations, the nominal inflation rate (yearly/monthly) is rising, mainly due to the energy price shock. Core inflation shows a monthly decline and a slight yearly increase. This data combination implies short-term inflation is slowing down, but long-term inflation remains sticky or under pressure. Nominal inflation is driven by the energy shock. Combined with recent expectations released by Federal Reserve officials, this aligns with no rate hike in the short term but retains the expectation of a rate hike in December. Previously, I mentioned that from a conspiracy theory perspective, assuming data can be manipulated, the Fed’s goal is simple: to raise rates once in September as an insurance measure, then adjust the next rate hike expectation to December through data, no hike in October, and observe economic data for two months. As the December meeting approaches, data will gradually weaken the possibility of a December hike. If actual inflation is controllable, the hike could be postponed to Q1 next year or even canceled altogether. #September CPI to be released tonight, Fed rate hike divergence awaits resolution. Of course, all premises depend on whether the CPI, PCE, and employment data for September and October support this. Currently, the PCE statistical methodology has changed, so the key is to watch the CPI. As long as the CPI is not excessive, data will gradually weaken before the December meeting, which is a healthy rhythm—neither overheating the market nor letting the market naturally expect rate hikes to self-limit liquidity. Back to the point, tonight看盘看到有点心累，山寨和主流又在悄悄换座位了 你有没有那种感觉，明明大盘没怎么动，手里的币却像被抽走了魂？ 这两天翻 $CORE 的数据，心里咯噔了一下。节点数如果继续掉，跌破 10 这个关口，风险就不是"回调"两个字能盖住的了，那基本等于软退出。但反过来看，节点稳在 16 没再往下，BTC 质押量还在涨，短期算是有块遮羞布。最怕的是第三种剧本：质押的 BTC 往外撤，节点也跟着减，两个坏消息撞一起，那就不是情绪问题了。 说回板块。最近 BTC 像那种稳稳坐在主位的姐姐，波动收窄，资金没急着走。ETH 呢，有点想表现但底气不足，跟涨不跟跌，偶尔硬一下又软回去。山寨这边就惨了，特别是那种靠节点叙事撑估值的项目，一旦数据松动，跑得比谁都快。$CORE 就是个典型，它的价格现在交易的其实不是技术，是"节点还稳不稳"这个预期。 市场在定价什么？我觉得是在定价"安全垫"的厚度。BTC 的持有者现在不太愿意把币扔去质押了，宁可放冷钱包。ETH 的质押流出也没那么夸张，但山寨质押的意愿在肉眼可见地降。这层传导很微妙：主流越稳，山寨越难抢到注意力，因为大家的风险偏好没消失，只是变挑剔了。 偏多的路径也最近市场有个挺有意思的现象。 大家都在盯着BTC涨跌，却忽略了一个可能影响行情的重要因素：原油价格。 说实话，这次CPI我更关注油价带来的通胀压力。 根据市场预测，美国9月整体CPI同比可能达到3.6%，而核心CPI预计在2.5%左右。 为什么两者差这么多？ 因为核心CPI剔除了食品和能源价格，而整体CPI会直接受到汽油等能源价格上涨的影响。 这就意味着，即使整体通胀上升，也不一定代表所有商品都在同步涨价。 但问题来了。 如果油价持续高位运行，运输和生产成本也可能慢慢上涨，最终传导到其他商品价格。 这才是美联储真正需要警惕的地方。 📊 再看看BTC为什么会受到影响。 逻辑其实不复杂： 油价上涨 → 通胀压力增加 → 降息预期减弱或加息预期增强 → 美债收益率可能走高 → BTC等风险资产承压。 当然，这不是必然发生的结果。 如果CPI低于市场预期，美债收益率回落，BTC反而可能获得喘息机会。 💡 我现在最关注两个数据： 一个是整体CPI有没有超出预期。 另一个是核心CPI有没有明显回升。 如果只是能源推动整体CPI上涨，而核心通胀保持稳定，市场未必会像想象中那么悲观。 我觉得今晚真#BofA survey reveals stock-bond divergence, BlackRock increases short-term bond allocation early $BTC
In the survey, 37% of funds are still in tech stocks, while the trigger for convertible bonds is set at 6%.
▪️ BofA October survey (10/2–10/8, 161 respondents): net overweight in tech rose from 30% to 37%
▪️ This is the highest level since August 2020; semiconductors have replaced AI concepts as the most crowded trade
▪️ 37% believe the 10-year US Treasury yield must reach 6.0% before switching from stocks to bonds
▪️ 19% think the current yield level is sufficient to shift allocation to bonds
▪️ BlackRock has shifted to increasing short-term bond allocation; Bridgewater's Dalio says the US Treasury bear market is far from over
▪️ Cash positions slightly increased, risk appetite cooled down
The divergence is not about who is right between stocks and bonds, but whether the survey measures position size or judgment.
The 37% net overweight records positions already bought, not bullishness level; the 37% who think bonds will only be switched at 6% measure a threshold not yet reached. Money is still in tech stocks, with the convertible bond trigger set at 6%—the stock-bond switch will not happen prematurely, only passively after yields move first.
The direction is to wait and see—if the 10Y yield does not hit 6%, positions continue; if it does, allocation shifts may be concentrated and released.#US-Russia Diesel Cooperation Faces Congressional Opposition, Ceasefire Conditions Still Divided $BTC
Whether this arrangement can be implemented depends not on the negotiation table, but on Congress.
▪️ The Trump administration plans to use Russian diesel exports to the US as leverage to push Russia back to negotiations
▪️ The arrangement has sparked bipartisan opposition in Congress, with disputes extending to legal authorization and ceasefire enforcement
▪️ The UK, Germany, Estonia, Sweden, and others have expressed disapproval
▪️ The UK Defense Secretary said this move is largely motivated by midterm election considerations
▪️ Trump announced on 10/11 that Russia and Ukraine have entered an "energy ceasefire" state, claiming both sides have agreed
▪️ Analysts believe short-term additional Russian diesel supply is limited and delivery capacity is uncertain
The disagreement is not about whether diesel prices will drop, but about how many approvals this arrangement must pass.
At the executive level, what can be signed is authorization; at the legislative level, it concerns the boundaries of sanction exemptions—bipartisan opposition targets the latter. Authorization can be reversed, exemptions require legislation. What determines whether this arrangement can become a long-term supply is not the statements at the negotiation table, but whether Congress is willing to provide a stable legal framework.
The direction is to wait and see — only a shift to long-term contracts or exemption legislation counts as increased supply; staying at authorization is merely a price disturbance.Tonight (October 14) at 20:30, the US September CPI will be released. Market expectations are already quite consistent: overall CPI year-on-year rising from 3.4% to 3.6%~3.7%, month-on-month about 0.6%; core CPI year-on-year steady at 2.4%, month-on-month about 0.22%, slowing compared to August's 0.29%.
The key point is: the overall inflation jump has already been priced in by the market because the driver is gasoline (September average price over $4.3 per gallon, a record for the month, with oil prices high due to the Iran conflict), which is considered a "one-time disturbance." So what really determines the direction of cryptocurrencies tonight is the core CPI, not the overall CPI. Don't go all in yet, wait and see tomorrow night's PPI data.ASML's Q3 revenue in euros was slightly below expectations, but slightly exceeded expectations in US dollars. This difference in measurement is not significant by itself; the key point is that EPS exceeded expectations by 20.8%, indicating improvements in cost control and a higher proportion of high-margin products. The quality of profit is better than the revenue figures.
This marks the third consecutive quarter of upward revisions to the outlook. The Q4 net sales guidance is €14.2 billion to €15.2 billion, while analysts previously expected around €13.97 billion. Such continuous upward revisions usually mean that management sees an extended order visibility.
Some customers have already started expressing purchase intentions for equipment in 2029 and 2030. The existing supply plans for 2027 and 2028 can no longer fully meet demand. Buyers willing to queue up in advance for years not yet arrived says more than a single quarter's performance.
Demand has outpaced capacity, so the constraint is no longer order volume but how many units can be produced. ASML's capacity ceiling is the upper limit for the entire downstream AI hardware expansion speed. With scheduling pushed to 2030, the barrier to entry is no longer just a matter of budget.
The official plan is to continue exploring the possibility of further capacity increases in 2028, building on the previous 30% expansion target. This signal should be viewed in conjunction with another fact: if the pace of expansion cannot keep up, some customers may not get the delivery times they want, and no matter how strong the demand, it will not translate into current period revenue.